EBS Workflow Mailer Failing To Send Email Notifications With java.net.ConnectException: Connection refused After Environment Configuration Change (Doc ID 1950783.1)

Last updated on AUGUST 08, 2016

Applies to:

Oracle Workflow - Version 12.1.3 to 12.1.3 [Release 12.1]
Information in this document applies to any platform.
During the course of investigation all attempted solutions below failed to resolve the issue. The 'connection refused' stubbornly remained.

1.Note 578578.1 What Is The Correct Value of Mailer Parameter "HTML Agent" In R12?
2.Verified profile option APPS_FRAMEWORK_AGENT
3. Verified profile option WF:workflow mailer framework web agent
4.Dropped and re-register the physical server

EXEC FND_CONC_CLONE.SETUP_CLEAN;
Commit;
5. Connection Refused Errors in Mailer Log After Applying ATG RUP4 (Doc ID 403431.1)
6. Java Notification Mailer Fails to Send Email Notifications from FormatterException with Framework Regions (Doc ID 339718.1)
7. Racing Condition May Throw Exception Creating New Poolable Object: oracle.apps.fnd.common.PoolException: Exception creating new Poolable object (Doc ID 1598680.1)
8. Workflow Agent Listeners Not Starting (Doc ID 796089.1)
9.Workflow Mailer Failing to Start with "Java.Net.Unknownhostexception" Error. (Doc ID 602651.1)

Symptoms

 Workflow Mailer is failing to send email notifications with:

oracle.apps.fnd.wf.mailer.NotificationFormatter$FormatterSAXException: Problem obtaining the HTML content -> oracle.apps.fnd.wf.common.HTTPClientException: Unable to invoke method HTTPClient.HTTPConnection.Get caused by: java.net.ConnectException: Connection refused

  
History:
A two server configuration for High Availability. One server is active, the other is idle.  They are clustered using VCS.
The system failed over to the idle server over the weekend. Using a BIG IP configuration to mask the two physical server names, and everything runs as expected, with this exception.

The environment is NOT SSL enabled. Nor it is using load balancing.

The file system and database are on a single server.  The RDBMS, App Tier, Concurrent managers all run on one server. All things are running fine, except, that the WF Mailer does not send email.

The key error messages in java mailer log files are:

a) oracle.apps.fnd.wf.mailer.NotificationFormatter$FormatterSAXException: Problem obtaining the HTML content -> oracle.apps.fnd.wf.common.HTTPClientException: Unable to invoke method HTTPClient.HTTPConnection.Get caused by: java.net.ConnectException: Connection refused

b) EXCEPTION:[jtf.cache.appsimpl.AppsCacheFrameworkConfig]:oracle.apps.fnd.common.AppsException: oracle.apps.fnd.common.PoolException: Exception creating new Poolable object.

c) java.lang.RuntimeException: oracle.apps.fnd.common.AppsException: java.sql.SQLException: The Network Adapter could not establish the connection

d) java.net.UnknownHostException: erpdbpcov1.fmr.com
at java.net.InetAddress.getAllByName0(InetAddress.java:1157)
at java.net.InetAddress.getAllByName(InetAddress.java:1083)
at java.net.InetAddress.getAllByName(InetAddress.java:1019)
at oracle.net.nt.TcpNTAdapter.connect(TcpNTAdapter.java:114)
at oracle.net.nt.ConnOption.connect(ConnOption.java:123)
at oracle.net.nt.ConnStrategy.execute(ConnStrategy.java:337)

e) ERROR:[jtf.cache.CacheManager]:Failed to load CacheComponent configuration

Changes

There is no change in java mailer configuration. Nor any change to email inbound or outbound servers.

The application uses virtual name in the URL.
The physical servers, LCOV55 and LCOV54 are both discovered in FND_NODE, but the target node shows the correct physical server name when running  this query:


select target_node from fnd_concurrent_queues where concurrent_queue_name like 'WFMLRSVC%';

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms